Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    names and related concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     are
    0.52
    :
    0.50
    5
    0.50
    -
    0.49
    4
    0.47
    8
    0.45
    ↵
    0.45
    ↵↵
    0.45
    по
    0.45
    те
    0.44
    POSITIVE LOGITS
    MatContext
    0.48
    RawContext
    0.45
    SXml
    0.45
    tbLabel
    0.45
     रोजिक
    0.45
    markUp
    0.44
     kho
    0.44
    SecurityMarks
    0.44
     ähm
    0.43
    តា
    0.43
    Act Density 0.000%

    No Known Activations